Analysis refers to the processing of text or data: Statistical and quantitative methods of analysis have brought close reading of texts (stylometrics and genre analysis, collation, comparison of versions for author attribution or usage patterns) into dialogue with distant reading (the crunching of large quantities of information across a corpus of textual data or its metadata).
Burdick et al. “1: Humanities to Digital Humanities,” in Digital_Humanities (Cambridge, MA: MIT Press, 2012), 3.
This excerpt from the chapter “Humanities to Digital Humanities” in the book Digital_Humanities particularly resonated with me, as I think that so-called “distant reading,” when combined with the process of close reading, is a very powerful tool for textual analysis. In my educational experience thus far, human-generated textual data remained in the realm of English and History classes, where we would perform what the author describes as close reading: genre analysis, style analysis, and frequent thematic analysis. Science classes, on the other hand, always focused on hard data: discrete data points where we would draw relationships between variables to support or reject our hypotheses. These realms never really crossed, which I think is the place where digital humanities comes into play and serves a critical role in applying the scientific methods of understanding to softer forms of data.
Analyzing humanities data, for example English language text corpora, serves a very important role and is something that is often overlooked. Utilizing statistical and linguistic analysis methods on such large scale data can help pick out information and trends that would otherwise be merely educated guesses. One application of these techniques that I could see, and that I know already have been done to some extent, is utilizing distant reading techniques to analyze news articles. Rather than picking out the individual statements said in one article, large scale trends can be drawn across a dataset of news articles, highlighting relevant factors like political partisianship, participation, and violence over time.
This term, I am particularly excited to work on developing my data visualization skills. I think that a well made graphic can effectively convey information much more efficiently than written conclusions, and I hope to be able to make use of these techniques to display analysis of my own. Additionally, I am taking CS 322: Natural Language Processing this term, and I am excited to possibly apply NLP techniques to this class to possibly do some distant reading of my own.
